~~NOTOC~~
======  Minute of meeting on 2019.03.06 ======
----

===== 1.  Attendees  =====
  * Alexander Harding (SEAKR) \\
  * bostromarna (?) \\
  * Bryan Fodness (Harris) \\
  * Christian M. Fuchs (Leiden University, Netherlands) \\
  * Dan Elftmann (Lockheed Martin) \\
  * Drew Boudreau (Trident Systems) \\
  * Eliot Glaser (Northrop Grumman) \\
  * Gary Swift (meeting moderator, Swift ERS) \\
  * Jim Devereaux (Xilinx FAE) \\    
  * Mark Handly ()\\
  * Michael Eller (Boeing) \\ 
  * Minal Sawant (Xilinx) \\ 
  * Patrick Fleming (Raytheon) \\
  * Peter Parsons (SEAKR) \\
  * Ron Smith (Northrop Grumman) \\
  * Sebastian Garcia (Slabs) \\
  * Stephen Thomas (Boeing) \\
  * Steve Pearl (General Dynamics) \\ 
  * Tesat (Tesat Spacecom, Airbus DS) team \\   
  * (2) non-identified call-in users \\


===== 2.  Topics covered =====

//NOTE: The following are very brief descriptions. Being this a Wiki, please complete/expand/correct them if you think we are missing interesting information (especially regarding anything important for future reference). 
//
//Paragraphs below are comments made by Gary (unless otherwise indicated).//

===== KU060 Upcoming tests =====

==== Interim (non-XRTC CTI) tests to come ====

Recent beam test schedule updates:
  * April @ TAMU: SEAKR to test KU060 MGTs (GTHs) with their own apparatus;\\
  * May: Boeing to test MGTs using a development/eval board (Gary mentioned VCU707 board, but this is a Virtex-7 based board, probably he was referring to a Virtex-US card).\\

Alex says SEAKR's plan is to test the GTH tiles on UltraScale (no protocol, just raw upsettability).\\
Dan comments that it would be interesting to test also protocols, specially Next Generation Space Interconnect Standard, NGSIS [1] (consortium developing an evolution of RapidIO std. together with VPX std.: SpaceVPX). Xilinx has an alliance partner (Praesum Communications) that has the IP core running on the KU060.\\
Praesum Communications is working with the high-perf. space computing group at Boeing. Talk with Jim Butler. \\
Steve (Boeing) will start also testing the basic GTHs, but he might be open to test the available NGSIS IP. \\
Dan had conversations about NGSIS with Richard Burger and Joe Marshall from BAE. One thing: trying to get that 32nd. full xver (the one moved around in the KU060 space part's package) back to the  NGSIS 32-port switch.\\

==== KU060 DUT board development status ====

Gary asks Boeing for updates on the schematic capture.\\
Michael Eller: Doing the V7-to-KU060 port of signals, changed some capacitors; there's a lot of different parts that we don't have in our component library.\\  


===== V7 Upcoming tests =====

April 2nd. Aerospace test slot: Would be useful to check-out the JCM-SMap handling of SEFIs using the V7 "straddle" board (with or without backplane). 
All XRTC tests going forward gonna use the JCM as ConfigMon, so there's a need to test this. 


Stakeholders: BAE, and now Raytheon with interest on testing MGTs (GTX & GTH). Patrick has booked UCD (protons) time for the week of April 2nd. \\
Patrick says they're developing on the backplane with DUT "straddle" cards, don't have ConfigMon set-up yet, gonna skip the BrainBox (BrainBox' ribbon cables not supported by straddle card).\\
Gary: It will be the first time for the backplane in the beam.\\

Jim Devereaux will check-out the GTH vs. GTX (GTHs available for testing in the 980T, How different would be w.r.t. GTXs regarding SEE susceptibility?).

New spin-off team meeting to come, regarding MGTs considerations for upcoming tests.

Gary: Using the straddle board, it turns out that our previous plan, pluging-in the Xilinx' XM105 board to work as a BrainBox I/F for FuncMon, is not possible. 
On the FuncMon end, we are thinking on using a serial port and pipe-out the FuncMon I/F in this way (instead of using 40-pin ribbon cables to the BrainBox). 
May be slower to get the data out of the target room, but we expect the overall speed to be appropriate.
We will try to take the GUI SW adapting it from the custom ribbon 40-pin DIO translation card that relies on WinXP and PCMCIA card.\\
BAE is also working on the FuncMon I/F, for now maintaining the BrainBox in there, but they have a plan to use high-speed serial signals.

Gary: Patrick's plan in terms of HW setup: 
  * Two straddle V7 cards pluggen on the backplane;\\
  * a (new) SMap-based JCM module plugged on the DUT end (scrubbing);\\
  * a (regular) JTAG-based JCM plugged on the FuncMon end (out of beam), freeing pins for use as serial output port.\\

    
==== V7 Full DUT card development status ==== 

No news from Neil (STS); we expect them to finish the board routing today. Preliminary delivery date: May 2nd.\\


===== Data analysis =====

Gary recalls the V5-QV clocking testing errata generated by Kevin Wray, as an outcome from V7 analysis. \\

Still trying to understand 7-Series BRAM remapping ("clobber") events, out of the 2017 data. \\ 








==== Round table ==== 

Christian Fuchs requests the MGT tests documentation for V-5, authored by Roberto Monreal. \\ 


[1] http://rapidio.org/wp-content/uploads/2014/10/RapidIO-NGSIS-Seminar-July-23-2014.pdf 

